151751-Najiba-Chargi

77 Diagnostics: cut-off values for low skeletal muscle mass INTRODUCTION Research on body composition in cancer patients, and in particular on skeletal muscle mass (SMM), has increasingly gained interest over the past several decades. Low SMM is often referred to as sarcopenia, although a more comprehensive definition of sarcopenia is the combination of low SMM and lowmuscle function. 1 Due to the unavailability of routinely per- formedmuscle function tests, most research in oncological patients focusses on radiologically assessed SMM, measured on routinely performed computed tomography (CT) or magnetic resonance imaging (MRI). Radiologically assessed low SMM at diagnosis has shown to predict adverse outcomes in a variety of cancer types and treatments. 2–6 For HNC, low SMM has shown to be a significant predictive factor for cisplatin dose-limiting toxicity 7 , the occurrence of a fistula after total laryngectomy 8 and flap-related complications in microvascular free flap head and neck reconstructive surgery. 9 It has also been shown that low SMM is prognostic for decreased survival in patients with HNC. 10–14 Several diagnostic imaging modalities can be used to quantify SMM such as magnetic reso- nance imaging (MRI), computed tomography (CT), bioimpedance analysis (BIA) and dual energy X-ray absorptiometry (DEXA). BIA and DEXA are confounded by alterations in hydration, edema and food intake. Therefore, its use in assessing body composition of patients with cancer is not favored. First research on body composition was performed by measurement of skeletal muscle area (SMA) on a single axial-slice at the level of the third lumbar vertebra (L3). 15,16 The skeletal muscle area (SMA) at the level of L3 is then normalized for height to calculate the lumbar skeletal muscle index (lumbar SMI), which is used as a proxy of whole body skeletal muscle mass. 16 Abdominal CT imaging is not routinely performed in patients with HNC and is often only available in patients with advanced disease and those at risk for distant metastasis. Measurements of SMM at the level of the third cervical vertebra (C3) have shown to correlate well with SMM measurements at the level of L3. 17 Therefore, in order to avoid selection bias (i.e. only patients with abdominal CT included) in research on SMM in HNC, measurement of SMA at the level of C3 is the preferred method. Measurement of SMA at the level of C3 con - sists of segmentation of both sternocleidomastoid muscles and the paravertebral muscles. If preferred, the SMA at the level of C3 can be converted to SMA at the level of L3 by using a previously published and validated prediction formula. 17 Accurate diagnosis of low SMM in clinical practice is impeded by heterogeneous cut-off values used to diagnose patients with low SMM. In oncological literature different cut-off values for lowSMM are used. Themost used cutoff values in the field of research on body composition are the ones defined by Prado et al. and Martin et al. 15,18 Prado et al. used optimum stratification analyses between muscle mass and mortality in a population of 250 obese (body mass index (BMI) ≥ 30 kg/m 2 ) patients with respiratory or gastro-intestinal malignancies and found cut- off values for low muscle mass to be 52.4 cm 2 /m 2 for men and 38.5 cm 2 /m 2 for women as the 5